Nanoscale topography, semiconductor polarity and surface functionalization: additive and cooperative effects on PC12 cell behavior

Abstract

This work compares the behavior of PC12 cells on planar and patterned III-nitride materials with nanostructured topographies. Three different materials' compositions containing N-polar and Ga-polar areas are studied: Al_0.8Ga_0.2N, Al_0.7Ga_0.3N, and GaN. Surface microscopy and spectroscopy, along with biological assays are used to understand the connection between nanoscopic features, polarity and surface functionalization. All materials are modified using a solution based approach to change their surface composition. The results demonstrate that altering the surface hydrophobicity can be used to generate additive effects with respect to protein adsorption in addition to the cooperative effects observed with respect to planes' polarity and topography. The work also details differences in the release of metal ions from clean and functionalized nanostructured III-polar and N-polar semiconductors in cell culture media, and their relationship to changes in cell response through quantification of cell viability and the production of reactive oxygen species. Our results demonstrate that nanoscale topography can be linked to additional parameters at the cell-semiconductor interface in order to understand and modulate PC12 cell behavior.

---

Introduction

Historically, advancements in electronics have relied on improving material characteristics responsible for properties such as carrier concentration control and their corresponding mobility that lead to better device performance in terms of speed and power. In contrast, advancements in bioelectronics in recent decades have embraced the integration of materials properties that enable favorable conditions for a specific bio-entity that is interfaced with an electronics' component.¹ For example, the incorporation of nanostructures as building blocks for sensing and recording modalities has resulted in bioelectronic devices for intracellular recording.² At present there is an enormous unmet need for more functional bioelectronic devices that can be mass produced in a manner similar to conventional semiconductor devices. Utilizing materials that are amenable to mass fabrication and enable modulation of multiple parameters with relevance to biointerfaces can provide a route to better bioelectronics.³ In particular, semiconductors that can be processed on wafer scale offer flexibility in bioelectronics device design as one can tune their electrical properties using high throughput localized processing such as doping. At the same time, one can utilize high yield and output processing to modulate characteristics such as surface chemistry and topography that are essential for specific behavior of bio-entities such as cells and biomolecules. Numerous reports have detailed the role of nanoscale topography on cell adhesion, growth and proliferation, and an equally prolific efforts have examined biomolecular adsorption on nanostructured surfaces.^4–9 Additional literature also exists to understand how nanotopography and mechanical properties can synergistically influence differentiation as well as disease progression.^10,11 However, there are no integrated efforts to date to understand the interplay between nanoscale topography and properties with relevance to semiconductor based electronics. Specifically, semiconductor polarity associated with piezoelectric effects and spontaneous polarization, a subject of many studies in the semiconductor community,¹² has not been studied in the context of cell behavior in the presence of variable nanoscale surface features and chemistry.

Polarity is an inherent property of materials with the wurtzite structure. As a result of it, group-III semiconductors (i.e. AlN, GaN, and InN) have a spontaneous polarization along the crystallographic c-axis or the [0001] direction, Fig. 1.¹³ Research on deposition methods has resulted in the ability to grow nitrides in the c-direction of its lattice with very distinct polar orientations, in the case of GaN: gallium polar (Ga-polar) and nitrogen polar (N-polar).¹⁴ The polarity orientation determines a number of properties such as ability to incorporate dopants as well as chemical stability and reactivity.¹⁵ The differences in reactivity of these surfaces with opposite polarities can lead to the fabrication of materials with distinct nanoscale topographies and lithographic features after etching.^16–19 In prior work we have demonstrated that Ga-polar and N-polar surfaces with variable topography can influence cell adhesion and growth.²⁰ However, there are no efforts in the literature to date to explore the role of the polarization orientation on the fate of cell behavior when other parameters of importance to cell behavior such as morphology and surface composition are altered. Recent work has pointed to the importance of surface polarity on cell proliferation and migration on ferroelectric lithium niobate.^21,22 Such results have prompted our interest to consider possible additive and cooperative effects on cell behavior when nanoscale topography, surface chemistry and planes' polarity can be modulated at a semiconductor interface.

Fig. 1 Schematic representation of the Ga-polar and N-polar surface of GaN, along with the surface functionalization strategy utilized in this work.

In this study we utilize Al_0.8Ga_0.2N, Al_0.7Ga_0.3N, and GaN with different polarities and lithographic patterns. We compare the behavior of PC12 cells on two different planar morphologies as well as lateral polarity structures composed of alternating III- and N-polar regions. In particular, we show that nanoscale features pre-determined by the semiconductor surface polarity play the most important role in the initial cell fate. We further modulate the interface by chemical functionalization and show that changes in surface hydrophobicity can be used to generate additive effects with respect to cell behavior in addition to the cooperative effects observed with respect to the polarity and surface topography. The work also examines differences in the stability of clean and functionalized nanostructured III- and N-polar oriented semiconductors in cell culture media, and their relationship to changes in cell response through the production of reactive oxygen species.

Results and discussion

We chose to use neurotypic cells for this work due to their relevance for bioelectronics applications.²³ In addition, the chosen PC12 cells are a widely accepted model system to evaluate nerve growth factor (NGF) induced neuronal differentiation under in vitro conditions.²⁴ The III-nitride material system presents a unique biocompatible surface that has been reported by different research groups as a suitable interface for both ex vivo biosensing^25,26 and in vitro studies with variable nanoscale moieties.^27,28 The III-nitride surfaces of five different material configurations were utilized in this study. We chose to examine both planar and patterned substrates. Two planar types of GaN were studied: GaN hillock and GaN step. The designations allow to differentiate between their distinct nanotopographies that result from the growth method employed during fabrication.²⁹ The patterned substrates contained alternating regions of III-polar and N-polar material, and have been referred to in the literature as lateral polarity structures (LPS).^30,31 Each polarity region on individual samples had the same width. Three different compositions of LPS were compared to the planar surfaces: GaN, Al_0.8Ga_0.2N, Al_0.7Ga_0.3N. All five types of samples were functionalized via an environmentally friendly benchtop technique. The functionalization is solution based and has been used by us³² and others.³³ After cleaning, we passivated the surface with an oxyhydroxide layer³⁴ by treatment with H₂O and H₂O₂, Fig. 1. The attached hydroxyl groups are then utilized for further attachment of a short chain hydrocarbon molecule, 4-chlorobutyric acid, Cl(CH₂)₃COOH. 4-Chlorobutyric acid was chosen mainly for its high solubility in water, and also because it contains a halogen with a known binding energy that can be tracked during surface spectroscopy studies. Using X-ray photoemission spectroscopy (XPS), covalent attachment onto the III-nitride surface was confirmed using the Cl 2p peak centered at 200 eV (see data in ESI†). Chemical functionalization was done to passivate the surface rather than introduce specific biomolecular features that can alter the propensity of cells to adhere better through molecular cues. In subsequent sections we compare the characteristics of the five types of samples before and after functionalization, and quantify the behavior of PC12 on these materials.

Nanoscale topography of planar vs. lateral polarity structures before and after functionalization

Representative AFM topography images are shown in Fig. 2 to illustrate the differences in the nanoscale features. Such differences contribute namely to the overall material hydrophobicity and roughness, which can all drastically impact cell viability and biocompatibility. The GaN hillock morphology is characterized with large spirals distributed throughout the surface, whereas the GaN step samples are uniform throughout and lack any distinguishing features. On the LPS samples, the polarities are labelled as III-polar surface (either Ga or Al) as the top, raised in topography stripe of the LPS, and the N-polar surface as the bottom, lower in topography stripe of the LPS. Additional AFM data is presented in the ESI.† There are significant differences in RMS values between areas of different polarities. In all samples the N-polar regions were always rougher.

Fig. 2 AFM topography images of (A) GaN hillock; (B) GaN step; and (C) GaN LPS.

By functionalizing the semiconductor surface with Cl(CH₂)₃COOH, the material hydrophobicity may be modulated to be more or less hydrophobic depending upon the initial morphology, Fig. 3. Such modulation is important in many biological applications as controlling the relative hydrophobicity is often a critical factor for cell attachment. In the case of GaN hillock, functionalization increased hydrophobicity from 76° to 99° thus changing its potential for cell attachment. Al_0.8Ga_0.2N also saw a significant change after functionalization, though in the opposite direction, from 98° to 63°, becoming more hydrophilic after functionalization. No significant differences were measured between the clean and functionalized surfaces within the other three samples, though the cleaned GaN hillock was measured to be more hydrophilic than the cleaned GaN step, cleaned GaN LPS, and cleaned Al_0.8Ga_0.2N. Using the topographical information provided by AFM, this logic follows as the GaN hillock structure shown in Fig. 2 has significantly different RMS values than GaN LPS III, GaN LPS N, Al_0.8Ga_0.2N LPS-N, and Al_0.7Ga_0.3N LPS-N. There is no significant difference in RMS between GaN step and GaN hillock, though the difference in hydrophobicity can be explained by examining the differences in the GaN hillock and GaN step topographies. The characteristically more heterogeneous GaN hillock surface has a low surface energy that allows greater spreading of water droplets as compared to the more homogeneous GaN step structure. Changes in hydrophobicities are expected to affect cell adhesion, however it is difficult to predict how specific cells will react to chemically functionalized nanostructured surfaces.³⁵ Generally, a number of researchers have concluded that the type of protein determines if it will adsorb more onto either hydrophobic or hydrophilic surfaces.³⁶ Since cells do not interact with the inorganic material directly but rather attach through an adsorbed protein mediated interface, we studied changes in roughness after each surface was soaked in cell media solution.

Fig. 3 Comparison of the hydrophobicities of the samples before and after functionalization.

We tabulated differences in RMS between samples that have been cleaned (clean) and functionalized with H₂O₂/Cl(CH₂)₃COOH (Func.), Fig. 4. The cleaned and functionalized samples were then soaked in cell culture media (DMEM) for 24 hours, and the RMS values were again measured and represented by clean-media, and Func. Media, respectively. We observed that the RMS is highly dependent upon the polarity of the semiconductor surface. For all three LPS structures GaN LPS, Al_0.8Ga_0.2N LPS, and Al_0.7Ga_0.3N LPS, the N-polar surfaces RMS were measured to be significantly higher than that of the III-polar surfaces. Conversely, there was no difference in RMS between GaN hillock and GaN step, though GaN LPS-N has a higher RMS than that of the GaN step and GaN hillock. GaN step and GaN hillock were also measured to have a lower RMS than that of GaN LPS-N and GaN LPS-III. It is easily observed that the functionalized surfaces soaked in DMEM of Al_0.7Ga_0.3N LPS-N and Al_0.7Ga_0.3N LPS-III are higher than that of the clean surfaces, and the cleaned surfaces soaked in DMEM. This is due to more protein accumulation on the functionalized surfaces than clean surfaces. A similar observation is made when looking at the data for the cleaned GaN step and functionalized GaN step. There were no differences between different conditions within GaN hillock, Al_0.8Ga_0.2N LPS-N and Al_0.8Ga_0.2N LPS-III. At the same time GaN LPS-III and GaN LPS-N both saw drastic increases in RMS values for clean surfaces that were soaked in media over those that were functionalized and soaked in media. This is likely due to the opposite effect of Al_0.7Ga_0.3N samples, whereby protein adsorption on the clean surfaces is greater than that of the functionalized surfaces. High RMS surfaces usually lead to greater hydrophobicity and thus can have an additive effect on cell adhesion trends. Prior studies with PC12 cells on Si₃N₄ have reported that when the roughness and the hydrophilicity (lower contact angle) are increased, the attachment of neurotypic cells is facilitated.³⁷ Other work with PC12 cells and oxide materials has identified an optimal combination of nanotopography and hydrophobicity for protein adsorption and subsequent cell adhesion at contact angles above 90° and pore size of 80 nm.³⁸ Our goal in this study is to identify a possible synergistic action among topography, surface chemistry and polarity of the semiconductor surface. With that in mind we note again that initial differences in the nanoscale topography of the clean LPS samples come as a result of the different surface energies arising from the different polar orientations. In addition, the observed changes in RMS values support the notion that the nature of the surface polarity (III-polar vs. N-polar), not just initial morphological differences, plays a role in the amount of proteins adsorbed from cell media onto the surface.

Fig. 4 RMS values before and after chemical functionalization as well as after soaking in cell culture media.

Cell behavior in the presence of planar vs. lateral polarity structures before and after functionalization

Cell culture studies were done over a period of 7 days with all sample types. At different time points we assessed cell viability, reactive oxygen species (ROS) production, and compared the number of cells attached per surface area. Viability was tracked using an alamarBlue® (AB) assay, Fig. 5, wherein AB was added to cells in culture at a 10% volume to the media volume and allowed to incubate for 4 hours at 37 °C. Within each day, there were no significant differences between any of the samples and the control well which contained no semiconductor. This suggests that all materials tested had little cytotoxic effect. Additionally, there was a significant increase in AB reduction between day 1, day 3, and day 7, for all conditions test suggesting cell growth continued throughout the experiment regardless of the material tested. Similarly, for each condition there is a significant increase between day 1 and day 7, though no significant increases were seen between day 1 and day 7. The results support prior observations that III-nitrides are biocompatible in planar and nanostructured morphologies.^39,40

Fig. 5 AB viability with each sample type at different time points.

Reactive oxygen species (ROS) were measured at the time of plating, and at days 1, 3, and day 7, using the cell permeant reagent 2′,7′-dichlorofluorescin diacetate (DCFH-DA), a dye that measures peroxyl, hydroxyl and other ROS activity within the cell. Prior to the measurement, the cells were seeded at 2.4 × 10⁵ cells per well. Each material condition was run in three replicates, and after the addition of 60 μM DCFDA dissolved in DMEM, the cells were allowed to incubate for 30 min. After incubation each media-DCDFA replicate was split into quadruplicates and fluorescence measurements were then taken. The results are shown in Fig. 6. Within day 1, clean GaN LPS produced more ROS species than that of all other samples including the functionalized GaN LPS. The well which contained no semiconductor and cleaned GaN step were significantly lower than that of cleaned Al_0.8Ga_0.2N LPS and the Func. GaN LPS. There was no difference between functionalized Al_xGa_1−xN compositions and functionalized GaN hillock and GaN step. Within day 3, there is no significant differences between clean GaN LPS and functionalized GaN LPS, though both are significantly larger than all other conditions. Additionally, there is no difference between the control well and the functionalized samples aside from functionalized GaN LPS. Day 7 produced less ROS species than day 3, and within day 7 there was no difference between clean and functionalized GaN LPS, though both were significantly larger than that of other clean samples and control well within day 7. The control well as well as the cleaned samples of Al_0.7Ga_0.3N LPS, GaN hillock, and GaN step were lower than that of all functionalized surfaces in day 7. Our results show no significant differences in viability, but variable amounts of ROS generated when PC12 cells are in contact with different materials. Prior work has noted that ROS release can be triggered by metal ion release and the interaction of cells with crystal defects on the surface of a materials.^41–43 Gallium and aluminum metal release will be discussed in the subsequent section. The data in Fig. 6 supports the notion that chemical functionalization passivates the surface thus minimizing the role of variable surface defects present on the different materials with respect to the production of ROS.

Fig. 6 Changes in the production of ROS at different time points in the presence of each sample.

All nitride samples are transparent and permit easy examination of cells via optical microscopy, Fig. 7. Across day 1 to day 3 for the entire sample set, there was a significant decrease in cells attached to semiconductors surfaces. A reason for this observation can be surface degradation (see next section). Though most samples showed no differences from days 1 and 3, there were some noteworthy examples including clean Al_0.7Ga_0.3N-III and clean GaN step. There was a clear decrease in cell attachment between clean GaN LPS-N and functionalized GaN LPS-N, suggesting that functionalization decreases cell attachment onto the semiconductor surface. We also observed some difference, not statistically significant, in the amount cells attached to clean III-polar and N-polar Al_0.8Ga_0.2N by day 3. Overall, no noteworthy differences in cell attachment were measured between III-polar and N-polar surfaces of the same material composition, suggesting that different plane polarities at most only indirectly affect cell adhesion. This is in agreement with work on lithium niobate that recorded that polarity does not influence cell adhesion and proliferation, but makes a large difference in cell migration trends.²²

Fig. 7 (A) Cells per mm² determined from optical microscopy images on each sample surface; (B) optical microscopy image showing cells attached onto a clean Al_0.7Ga_0.3N LPS substrate.

Stability of planar vs. lateral polarity structures in cell media before and after chemical functionalization

For the duration of the in vitro studies we removed aliquots of cell media during specific time points. The media was analyzed by Inductively Coupled Plasma Mass Spectrometry (ICP-MS) for the total amount of aluminum and gallium present, Fig. 8. We focus on the aluminum release first, Fig. 8A. One can see measurable amounts of Al in the sample which contained no semiconductor, this is due to the trace amounts of Al present in the media solution that was used. The amount of Al allowed in drinking water based on EPA standards is currently set at 0.05 to 0.2 mg L⁻¹.⁴⁴ The only samples which showed significantly larger Al leachate concentration than the control well are the cleaned and functionalized Al_xGa_1−xN samples. In several Al_xGa_1−xN samples on day 7, there was a sharp increase in Al concentration. Within the cleaned samples of Al_0.8Ga_0.2N, there was a sharp increase from on day 1 to on day 7. Within the functionalized samples of Al_0.7Ga_0.3N, there was a sharp increase from on day 1 to on day 7. Very different observations were made for Ga leaching, Fig. 8B. Cleaned GaN step, cleaned Al_0.7Ga_0.3N, and functionalized Al_0.7Ga_0.3N all had increased amounts of Ga leaching between day 7 and days 1 and 3. Cleaned and functionalized GaN LPS samples saw significant increases in Ga leaching between day 1, day 3 and day 7. All other samples (clean and functionalized Al_0.8Ga_0.2N, GaN hillock, and functionalized GaN step) saw no significant differences between each day. The high leaching content of the GaN LPS samples is likely due to these samples having the highest % Ga content, while half of their surface areas contained N-polar stripes. The Ga release data correlates with the ROS data discussed above. Prior work with metal ions and PC12 cells has noted that they can induce apoptosis of PC12 cells by an increased generation of ROS species.⁴⁵ The samples that resulted in the highest amount of Ga leaching caused the greatest increase in ROS production. Stability, and therefore release of metal ions in solution, is directly related to the polarity of the surface. N-polar surfaces are less stable in solution compared to III-polar surfaces.⁴⁶ Therefore, polarity via its direct connection to the surface stability, does play a role in cell response through a metal dependent increase in ROS production. We note that the ROS production variability for each sample type is not significant enough over the duration of our experiments as to result in changes in PC12 cell viability. Our data indicates that surfaces with distinct nanoscale topography can lead to variable ROS production. The data we report in this work suggests that the ROS production from a nanoparticle⁴⁷ vs. ROS production from a nanostructured surface is through a similar but distinctly different mechanism. The instability of the N-polar surface in solution leads to release of Ga ions, but also results in changes in the amount of surface adsorbed proteins which mediate the cell-surface interactions and subsequent cellular behavior.

Fig. 8 ICP-MS analysis for the amount of (A) Al leached and (B) Ga leached during the cell culture experiments with all types of samples.

Conclusions

In this work we show that biocompatible III-nitride semiconductor materials with nanoscale topographies have additional characteristics that permit modulation of neurotypic cells behavior at the cell-surface interface. The semiconductor polarity does not have an effect on cell adhesion but is directly related to the stability in water solutions. Thus polarity is linked to changes in the production of reactive oxygen species through the release of group III metals in solution and changes in surface morphology available for protein adsorption which facilitates cell adhesion and succeeding cellular behavior. Chemical functionalization leads to changes in initial protein adsorption but results in smaller changes in cell adhesion over time.

Methods

Materials

GaN hillock and GaN step were both grown through metal–organic chemical vapor deposition and diced into 3 × 3 mm wafers. Al_0.7Ga_0.3N, Al_0.8Ga_0.2N, and GaN LPS wafers were produced through a sequential patterning, growth and etching^16,17 and diced into 5 × 5 mm wafers. 4-Chlorobutyric acid and 2′,7′-dichlorofluorescin diacetate were obtained from Sigma-Aldrich and used as received. Additionally, PC12 cells were obtained from Sigma-Aldrich. The sixth passage of PC12 cells were used in this study. AlamarBlue® and Dulbecco's Modified Eagle Medium (DMEM) was obtained from Thermo Fisher and used as received.

Contact angle

2 μL droplets of DI water were deposited with a Ramé-hart automated dispensing system and were imaged using a Ramé-hart Model 200 F4 series standard goniometer. Droplet images were analyzed with DropImage Standard v2.4.

Cell culture and cellular assays

PC12 cells were cultured on collagen-coated 24-well plates at a cell density of 2.4 × 10⁵ cells per well. Prior protocols⁴⁸ and supplier guidelines were followed. A Tecan GENios microplate reader with a 485 nm excitation filter and a 535 nm emission filter was used to measure fluorescence for the ROS assay. For the AB assay, 570 and 600 nm filters were used to measure the absorbance of each sample. Cells were incubated at 37 °C in a 5% CO₂ environment for the duration of the experiment.

X-ray photoelectron spectroscopy

XPS data were obtained with a Kratos Analytical Axis Ultra XPS. A survey scan was collected for each of the samples with a pass energy of 160 eV. High resolution regional scans of C 1s, O 1s, Ga 2p, Ga 3d, N 1s, and Cl 2p were obtained with a pass energy of 20 eV. Casa XPS v2.3.12.8 was utilized to fit peaks and determine atomic percentages of each sample from the survey scans. Background subtraction was done using a Shirley approximation and all peaks were calibrated by setting the adventitious carbon C 1s peak to 284.8 eV.

Atomic force microscopy

An Asylum Research Cypher S AFM was used to take three random scans containing height, phase, and amplitude information obtained via tapping mode in air at room temperature. A scanning frequency of 0.5 Hz was used to image a 3.0 μm × 3.0 μm area. Soft Si cantilevers obtained from Asylum Research, (f = 70 kHz, k = 2 N m⁻¹) were used for this experiment. The images were processed and the root mean square (RMS) roughness values was calculated through Igor Pro (v13.01.68) and Gwyddion (v2.42).

Statistical analysis

All statistical analysis was conducted using a significance level of 0.05, and performed with one-way and two-way ANOVA tests using OriginPro 2016 (v9.3).

Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

We thank Army Research Office under W911NF-15-1-0375 for support of this work. Partial financial support from NSF (DMR-1312582), is greatly appreciated. The work was performed in part at the Environmental and Agricultural Testing Services laboratory (EATS), Department of Crop and Soil Sciences, at North Carolina State University.

References

1. M. Birkholz, A. Mai, C. Wenger, C. Meliani and R. Scholz, Technology modules from micro- and nano-electronics for the life sciences, Wiley Interdiscip. Rev.: Nanomed. Nanobiotechnol., 2016, 8, 355–377.
2. A. Q. Zhang and C. M. Lieber, Nano-Bioelectronics, Chem. Rev., 2016, 116, 215–257.
3. S. Choi, H. Lee, R. Ghaffari, T. Hyeon and D. H. Kim, Recent Advances in Flexible and Stretchable Bio-Electronic Devices Integrated with Nanomaterials, Adv. Mater., 2016, 28, 4203–4218.
4. A. T. Nguyen, S. R. Sathe and E. K. F. Yim, From nano to micro: topographical scale and its impact on cell adhesion, morphology and contact guidance, J. Phys.: Condens. Matter, 2016, 28, 16.
5. P. Roach, T. Parker, N. Gadegaard and M. R. Alexander, Surface strategies for control of neuronal cell adhesion: A review, Surf. Sci. Rep., 2010, 65, 145–173.
6. D. Hoffman-Kim, J. A. Mitchel and R. V. Bellamkonda, Topography, Cell Response, and Nerve Regeneration, in Annual Review of Biomedical Engineering, ed. M. L. Yarmush, J. S. Duncan and M. L. Gray, Palo, Alto, 2010, vol. 12, pp. 203–231.
7. M. S. Lord, M. Foss and F. Besenbacher, Influence of nanoscale surface topography on protein adsorption and cellular response, Nano Today, 2010, 5, 66–78.
8. G. J. Hickman, D. J. Boocock, A. G. Pockley and C. C. Perry, The Importance and Clinical Relevance of Surfaces in Tissue Culture, ACS Biomater. Sci. Eng., 2016, 2, 152–164.
9. E. Ngandu Mpoyi, M. Cantini, P. M. Reynolds, N. Gadegaard, M. J. Dalby and M. Salmerón-Sánchez, Protein Adsorption as a Key Mediator in the Nanotopographical Control of Cell Behavior, ACS Nano, 2016, 10, 6638–6647.
10. D.-H. Kim, P. P. Provenzano, C. L. Smith and A. Levchenko, Matrix nanotopography as a regulator of cell function, J. Cell Biol., 2012, 197, 351.
11. C. Y. Tay, C. G. Koh, N. S. Tan, D. T. Leong and L. P. Tan, Mechanoregulation of stem cell fate via micro-/nano-scale manipulation for regenerative medicine, Nanomedicine, 2013, 8, 623–638.
12. S. Keller, H. R. Li, M. Laurent, Y. L. Hu, N. Pfaff, J. Lu, D. F. Brown, N. A. Fichtenbaum, J. S. Speck, S. P. DenBaars and U. K. Mishra, Recent progress in metal-organic chemical vapor deposition of (000(1)over-bar) N-polar group-III nitrides, Semicond. Sci. Technol., 2014, 29, 47.
13. C. E. Dreyer, A. Janotti and C. G. Van de Walle, Absolute surface energies of polar and nonpolar planes of GaN, Phys. Rev. B: Condens. Matter Mater. Phys., 2014, 89, 4.
14. J. K. Hite, N. D. Bassim, M. E. Twigg, M. A. Mastro, F. J. Kub and C. R. Eddy, GaN vertical and lateral polarity heterostructures on GaN substrates, J. Cryst. Growth, 2011, 332, 43–47.
15. O. Romanyuk, P. Jiricek, T. Paskova and I. Bartos, Polarity of GaN with polar {0001} and semipolar {10(1)over-bar1}, {20(2)over-bar1}, {11(2)over-bar2} orientations by x-ray photoelectron diffraction, J. Mater. Res., 2015, 30, 2881–2892.
16. M. P. Hoffmann, M. Gerhold, R. Kirste, A. Rice, C. R. Akouala, J. Q. Q. Xie, S. Mita, R. Collazo and Z. Sitar, Fabrication and Characterization of Lateral Polar GaN Structures for Second Harmonic Generation, in Quantum Sensing and Nanophotonic Devices X, ed. M. Razeghi, E. Tournie and G. J. Brown, Proceedings of SPIE, Spie-Int Soc Optical Engineering, Bellingham, 2013, vol. 8631.
17. M. P. Hoffmann, R. Kirste, S. Mita, W. Guo, J. Tweedie, M. Bobea, I. Bryan, Z. Bryan, M. Gerhold, R. Collazo and Z. Sitar, Growth and characterization of AlxGa1-xN lateral polarity structures, Phys. Status Solidi A, 2015, 212, 1039–1042.
18. R. Kirste, S. Mita, M. P. Hoffmann, L. Hussey, W. Guo, I. Bryan, Z. Bryan, J. Tweedie, M. Gerhold, A. Hoffmann, R. Collazo and Z. Sitar, Properties of AlN based lateral polarity structures, in Physica Status Solidi C: Current Topics in Solid State Physics, ed. S. Hildebrandt, Physica Status Solidi C-Current Topics in Solid State Physics 2; Wiley-V C H Verlag Gmbh, Weinheim, 2014, vol. 11, no 2, pp. 261–264.
19. R. Kirste, S. Mita, L. Hussey, M. P. Hoffmann, W. Guo, I. Bryan, Z. Bryan, J. Tweedie, J. Xie, M. Gerhold, R. Collazo and Z. Sitar, Polarity control and growth of lateral polarity structures in AlN, Appl. Phys. Lett., 2013, 102, 181913.
20. L. E. Bain, R. Kirste, C. A. Johnson, H. T. Ghashghaei, R. Collazo and A. Ivanisevic, Neurotypic cell attachment and growth on III-nitride lateral polarity structures, Mater. Sci. Eng., C, 2016, 58, 1194–1198.
21. N. C. Carville, L. Collins, M. Manzo, K. Gallo, B. I. Lukasz, K. K. McKayed, J. C. Simpson and B. J. Rodriguez, Biocompatibility of ferroelectric lithium niobate and the influence of polarization charge on osteoblast proliferation and function, J. Biomed. Mater. Res., Part A, 2015, 103, 2540–2548.
22. V. Marchesano, O. Gennari, L. Mecozzi, S. Grilli and P. Ferraro, Effects of Lithium Niobate Polarization on Cell Adhesion and Morphology, ACS Appl. Mater. Interfaces, 2015, 7, 18113–18119.
23. H. W. Liu, W. C. Huang, C. S. Chiang, S. H. Hu, C. H. Liao, Y. Y. Chen and S. Y. Chen, Arrayed rGOSH/PMASH Microcapsule Platform Integrating Surface Topography, Chemical Cues, and Electrical Stimulation for Three-Dimensional Neuron-Like Cell Growth and Neurite Sprouting, Adv. Funct. Mater., 2014, 24, 3715–3724.
24. S. Meucci, I. Tonazzini, F. Beltram and M. Cecchini, Biocompatible noisy nanotopographies with specific directionality for controlled anisotropic cell cultures, Soft Matter, 2012, 8, 1109–1119.
25. S. J. Pearton, B. S. Kang, S. K. Kim, F. Ren, B. P. Gila, C. R. Abernathy, J. S. Lin and S. N. G. Chu, GaN-based diodes and transistors for chemical, gas, biological and pressure sensing, J. Phys.: Condens. Matter, 2004, 16, R961–R994.
26. F. Ren and S. J. Pearton, Sensors using AlGaN/GaN based high electron mobility transistor for environmental and bio-applications, in Physica Status Solidi C: Current Topics in Solid State Physics, ed. M. Walther, Physica Status Solidi C-Current Topics in Solid State Physics 2, Wiley-V C H Verlag Gmbh, Weinheim, 2012, vol. 9, no 2.
27. J. Li, C. Qi, Z. Lian, Q. Han, X. Wang, S. Cai, R. Yang and C. Wang, Cell-Capture and Release Platform Based on Peptide-Aptamer-Modified Nanowires, ACS Appl. Mater. Interfaces, 2016, 8, 2511–2516.
28. G. M. Muntze, E. Pouokam, J. Steidle, W. Schafer, A. Sasse, K. Roth, M. Diener and M. Eickhoff, In situ monitoring of myenteric neuron activity using acetylcholinesterase-modified AlGaN/GaN solution-gate field-effect transistors, Biosens. Bioelectron., 2016, 77, 1048–1054.
29. R. F. Davis, S. M. Bishop, S. Mita, R. Collazo, Z. J. Reitmeier and Z. Sitar, Epitaxial growth of gallium nitride, in Perspectives on Inorganic, Organic, and Biological Crystal Growth: From Fundamentals to Applications, ed. M. Skowronski, J. J. DeYoreo and C. A. Wang, AIP Conference Proceedings, Amer Inst Physics, Melville, 2007, vol. 916, pp. 520–540.
30. D. Alden, W. Guo, R. Kirste, F. Kaess, I. Bryan, T. Troha, A. Bagal, P. Reddy, L. H. Hernandez-Balderrama, A. Franke, S. Mita, C. H. Chang, A. Hoffmann, M. Zgonik, R. Collazo and Z. Sitar, Fabrication and structural properties of AlN submicron periodic lateral polar structures and waveguides for UV-C applications, Appl. Phys. Lett., 2016, 108, 5.
31. S. Mita, R. Collazo and Z. Sitar, Fabrication of a GaN lateral polarity junction by metalorganic chemical vapor deposition, J. Cryst. Growth, 2009, 311, 3044–3048.
32. B. L. Pearce, S. J. Wilkins, T. Paskova and A. Ivanisevic, A review of in situ surface functionalization of gallium nitride via beaker wet chemistry, J. Mater. Res., 2015, 30, 2859–2870.
33. M. Eckshtain-Levi, E. Capua, Y. Paltiel and R. Naaman, Hybrid Sensor Based on AlGaN/GaN Molecular Controlled Device, ACS Sens., 2016, 1, 185–189.
34. S. J. Wilkins, T. Paskova, C. L. Reynolds and A. Ivanisevic, Comparison of the Stability of Functionalized GaN and GaP, ChemPhysChem, 2015, 16, 1687–1694.
35. M. M. Gentleman and E. Gentleman, The role of surface free energy in osteoblast-biomaterial interactions, Int. Mater. Rev., 2014, 59, 417–429.
36. S. P. Pujari, L. Scheres, A. T. M. Marcelis and H. Zuilhof, Covalent Surface Modification of Oxide Surfaces, Angew. Chem., Int. Ed., 2014, 53, 6322–6356.
37. J. J. M. Benavente, H. Mogami, T. Sakurai and K. Sawada, Evaluation of Silicon Nitride as a Substrate for Culture of PC12 Cells: An Interfacial Model for Functional Studies in Neurons, PLoS One, 2014, 9, 13.
38. H. J. Wang, Y. Y. Sun, Y. Cao, K. Wang, L. Yang, Y. D. Zhang and Z. Zheng, Is there an optimal topographical surface in nano-scale affecting protein adsorption and cell behaviors? Part II, J. Nanopart. Res., 2012, 14, 10.
39. L. E. Bain, R. Collazo, S. H. Hsu, N. P. Latham, M. J. Manfra and A. Ivanisevic, Surface topography and chemistry shape cellular behavior on wide band-gap semiconductors, Acta Biomater., 2014, 10, 2455–2462.
40. J. Y. Li, Q. S. Han, Y. Zhang, W. Zhang, M. D. Dong, F. Besenbacher, R. Yang and C. Wang, Optical Regulation of Protein Adsorption and Cell Adhesion by Photoresponsive GaN Nanowires, ACS Appl. Mater. Interfaces, 2013, 5, 9816–9822.
41. J. Lovrić, S. J. Cho, F. M. Winnik and D. Maysinger, Unmodified Cadmium Telluride Quantum Dots Induce Reactive Oxygen Species Formation Leading to Multiple Organelle Damage and Cell Death, Chem. Biol., 2005, 12, 1227–1234.
42. H. L. Karlsson, P. Cronholm, J. Gustafsson and L. Möller, Copper Oxide Nanoparticles Are Highly Toxic: A Comparison between Metal Oxide Nanoparticles and Carbon Nanotubes, Chem. Res. Toxicol., 2008, 21, 1726–1732.
43. M. Deng, Z. Huang, Y. Zou, G. Yin, J. Liu and J. Gu, Fabrication and neuron cytocompatibility of iron oxide nanoparticles coated with silk-fibroin peptides, Colloids Surf., B, 2014, 116, 465–471.
44. H. Galalgorchev, Who guidelines for drinking-water quality, in Iwsa Specialized Conference on Quality Aspects of Water Supply, ed. F. Fiessinger, Water Supply: The Review Journal of the International Water Supply Association 3-4, Blackwell Scientific Publ, Oxford, 1992, vol. 11, pp. 1–16.
45. R. Z. Lin, X. H. Chen, W. M. Li, Y. F. Han, P. Q. Liu and R. B. Pi, Exposure to metal ions regulates mRNA levels of APP and BACE1 in PC12 cells: Blockage by curcumin, Neurosci. Lett., 2008, 440, 344–347.
46. C. M. Foster, R. Collazo, Z. Sitar and A. Ivanisevic, Aqueous Stability of Ga- and N-Polar Gallium Nitride, Langmuir, 2013, 29, 216–220.
47. C. Y. Tay, M. I. Setyawati, J. Xie, W. J. Parak and D. T. Leong, Back to Basics: Exploiting the Innate Physico-chemical Characteristics of Nanomaterials for Biomedical Applications, Adv. Funct. Mater., 2014, 24, 5936–5955.
48. S. A. Jewett, M. S. Makowski, B. Andrews, M. J. Manfra and A. Ivanisevic, Gallium nitride is biocompatible and non-toxic before and after functionalization with peptides, Acta Biomater., 2012, 8, 728–733.

---

Footnote

† Electronic supplementary information (ESI) available: Additional microscopy and spectroscopy characterization. See DOI: 10.1039/c6ra21936e

---